Suggestions:

• Learning and Transfer of Modulated Locomotor Controllers –  https://arxiv.org/abs/1610.05182
• Benchmarking Deep Reinforcement Learning for Continuous Control  –
• Weight Uncertainty in Neural Networks – 
• Multiple Object Recognition with Visual Attention –  http://arxiv.org/abs/1412.7755
• Hierarchical Relative Entropy Policy Search –  http://jmlr.org/papers/v17/15-188.html
• Probabilistic inference for determining options in reinforcement learning –  http://link.springer.com/article/10.1007/s10994-016-5580-x